CN105760464A - Method for automatically displaying inferences and computing device - Google Patents

Method for automatically displaying inferences and computing device Download PDF

Info

Publication number
CN105760464A
CN105760464A CN201610082319.2A CN201610082319A CN105760464A CN 105760464 A CN105760464 A CN 105760464A CN 201610082319 A CN201610082319 A CN 201610082319A CN 105760464 A CN105760464 A CN 105760464A
Authority
CN
China
Prior art keywords
text
deduction
stream
word
term
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201610082319.2A
Other languages
Chinese (zh)
Inventor
埃泽齐埃尔·J·J·克鲁格里克
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Empire Technology Development LLC
Original Assignee
Empire Technology Development LLC
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Empire Technology Development LLC filed Critical Empire Technology Development LLC
Publication of CN105760464A publication Critical patent/CN105760464A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/903Querying
    • G06F16/9035Filtering based on additional data, e.g. user or group profiles
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9535Search customisation based on user profiles and personalisation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/951Indexing; Web crawling techniques
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/957Browsing optimisation, e.g. caching or content distillation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00Computing arrangements using knowledge-based models
    • G06N5/04Inference or reasoning models
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10LSPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00Speech recognition
    • G10L15/26Speech to text systems

Abstract

A method for automatically displaying inferences and a computing device. A method for automatically displaying inferences through a computing device comprises: receiving streaming audio data or streaming video data; when receiving the streaming audio data or the streaming video data, basically playing the received streaming audio data or streaming video data in real time; when playing the received streaming audio data or streaming video data, displaying streaming text corresponding to voice in the received streaming audio data or streaming video data; identifying inferences in the streaming text; visually indicating the identified inferences in the streaming text, wherein visually indicating the identified inferences in the streaming text comprises highlighting or bracketing the inferences identified in a text window including the streaming text; and displaying web search information correlated with the identifed inferences together with the identifed inferences.

Description

Automatically method and the equipment of calculating of term is inferred in display
The application be Application No. 201180004870.0, international application no be PCT/US2011/027390, filing date On March 7th, 2011, the divisional application of application for a patent for invention of invention entitled " virtual assistance inferred based on search engine ".
Cross-Reference to Related Applications
This application claims entitled " the SEARCH ENGINE INFERENCE BASED submitted on March 16th, 2010 VIRTUAL ASSISTANCE " the priority of U. S. application 12/724,660 (its attorney docket is MTCW001901).
Technical field
The present invention relates to the virtual assistance inferred based on search engine.
Background technology
Unless otherwise noted herein, the existing skill of otherwise method not claims hereof described in this section Art, and it is not recognized as prior art because it is included by this section.
When listening to or participate in session, speech or meeting, when presenting main topic of discussion or topic with begging for of being presented Additional information is received the most helpful during the theme of opinion or the relevant ad-hoc communication session of topic.But, in the session phase Between carry out manual web search and word for word analyzing and be typically inconvenient, and speed is very slow, can shift the notice of listener, Also listener and other people notice can be made away from current discussion.
Summary of the invention
Present disclosure describes the method for the virtual assistance inferred based on search engine, computer-readable medium and set Standby.Some exemplary methods may include that and automatically determine deduction, including receiving the input as the equipment of calculating of the stream text;Work as reception During to described text, based on the web search of term in described text, determine the deduction relevant with the theme of described text;And Automatically described deduction is shown.Determine and infer the result that may include that analysis web search, to determine the multiple of web search results Hit and topic.When the quantity of the hit returned is under predeterminated level, it is believed that the result of web search be enough to really The topic of vertical web search results, to determine deduction.When the quantity of hit is on predeterminated level, it is believed that web search Result be inadequate, it is possible to use the combination of the term that is gradually increased in text performs the web search added.Receive conduct The text of input can include the text produced by speech-to-text converter, can will infer that being supplied to speech-to-text turns Parallel operation, with the voice signal that auxiliary decoding is follow-up.Some example computer readable mediums include being configured to perform to be retouched here The computer executable instructions of the method stated, some example apparatus can include processor and the memory coupled with processor, Described memory has computer executable instructions, and when performing described computer executable instructions, configuration processor performs this Method described by.
Above summary of the invention is merely illustrative, and is by no means limitative.Except above-mentioned exemplary each scheme, respectively Outside embodiment and each feature, referring to the drawings and the following detailed description, it will be clear that other schemes, other embodiments and other spies Levy.
Accompanying drawing explanation
According to following description and claims, in conjunction with accompanying drawing, the foregoing and other feature of the disclosure will be apparent from. These accompanying drawings illustrate only some examples according to the disclosure, and is therefore not considered as limiting disclosure scope.Pass through Accompanying drawing is used to describe the disclosure in detail with extra feature and details, wherein:
Fig. 1 shows the flow chart of the instantiation procedure for the virtual assistance inferred based on search engine;
Fig. 2 A show according to based on search engine infer virtual assistance, for from as calculate equipment input The word received determines the flow chart of the instantiation procedure of deduction;
Fig. 2 B show according to based on search engine infer virtual assistance, for from as calculate equipment input The word received determines the flow chart of the example alternative procedure of deduction;
The virtual assistance that Fig. 3 is based on the communication of real-time reception in mobile device, basis is inferred based on search engine, The diagram of the example mobile device that display example is inferred;
Fig. 4 is the diagram of example networked computing environment, wherein can realize virtual assistance based on search engine deduction Aspect;
Fig. 5 shows the schematic diagram of the computer program for the virtual assistance inferred based on search engine;
Fig. 6 is the block diagram of the Example Computing Device that can realize virtual assistance based on search engine deduction thereon, entirely Portion is configured according at least some embodiment as described herein.
Detailed description of the invention
In the following detailed description, with reference to the accompanying drawing as the part described in detail.In the accompanying drawings, simileys leads to Often represent like, unless context indicates separately.Showing described in detailed description of the invention part, drawings and claims Example embodiment is not restrictive.Without departing from the case of the spirit or scope of this presented theme, it is possible to use Other embodiments, and other changes can be carried out.Should be appreciated that record and in the accompanying drawing disclosure of diagram general at this Each scheme can arrange according to multiple different configurations the disclosedest at this, replace, combine and design, and constitutes these public affairs The part opened.
As here will described in, relevant the setting of virtual assistance that more particularly, the disclosure relates to infer based on search engine Standby, method and computer program.
Technology as described herein is usually directed to based on the system inferred.Example embodiment can propose to draw with based on search Hold up the relevant equipment of the virtual assistance of deduction, method and computer program.One exemplary method may include that calculating equipment is suitable to Reception text is as input, and computer processor is arranged to based on the one or more terms in text one or many Individual web search determines that at least one relevant with the theme of text is inferred.Then, once it is determined that infer, can automatically show Infer.
Present disclosure recognize that and understand, traditional shows based on the system inferred, can derive additional information from discuss A kind of mode be the technology using deduction for having been discussed.But, current system based on deduction is typically The existing document designeding to be used under static environment.This system is not easy to the communication being applied to occur in real time, not Develop in time, or provide implication and the conclusion of renewal along with word and the change of context.
Fig. 1 show arrange according at least some embodiment described herein for inferring based on search engine The flow chart of the instantiation procedure 100 of virtual assistance.In shown example, can be described as processing step, feature operation, thing The process 100 of part and/or behavior etc. and other process as described herein, various functional block or behavior can pass through hardware, soft Part and/or firmware perform.Those skilled in the art according to the disclosure it will be appreciated that Fig. 1 can be put into practice in various embodiments The various modification of shown functional block.Such as, although process as shown in Figure 1 100 includes one specific order of piece or behavior, But theme required for protection need not be confined to any particular order by the order that these blocks or behavior are presented.Similarly, In the case of the scope without departing from theme required for protection, it is possible to use the behavior between two parties not shown in Fig. 1 and/or Fig. 1 Not shown in adjunctive behavior, and/or some behaviors shown in Fig. 1 can be removed.Process 100 can include operating 101,103 And/or one or more in 105, wherein can start to process at block 101.
At block 101, " receiving word or text chunk from speech-to-text engine ", word can be received or text chunk sets as calculating Standby input.For example, it is possible in unidirectional or double-directional speech or textcommunication, video conference, live chat session, stream audio or regard Frequently, during the playback during session or on the equipment of pre-recorded video or audio frequency etc., this word is captured by calculating equipment Or text chunk.In the case of first with phonetic incepting or generation communication, perform on the computing device (or receive at the equipment of calculating Performed before voice signal) real-time voice convert speech into text to text-processing, thus here described by base In the virtual assistance that search engine is inferred, each word of communication can process as text.Various currently available speeches Identify and speech-to-text application and system can be used for providing applicable speech-to-text session whole or in part, such as by Dragon System and Nuance Communications company provide those.Example Computing Device can include platform Formula computer, notebook, mobile computing device, smart phone, personal digital assistant etc..Can be block after block 101 103。
At block 103, " determining deduction in real time ", according to real-time constraint (or near real-time constraint), based on word one or more The web search of contamination in web search or text chunk, by calculating equipment determine from word or text chunk 103 1 or Multiple deductions.Deduction can also be based on by previously received word or during identical communication session or previous communication session The former network search that the word section that receives is constituted, and based on the text chunk and/or previously and follow-up text at communication session The frequency that in section, specific word or phrase occur.Show in Figures 2 A and 2 B use this web search to determine what these inferred shows Example process.Can be block 105 after block 103.
At block 105, " display in real time is inferred for the highest (top) of word/text chunk ", on the computing device, in real time (or connect Near real-time) show 105 for the word received and/or the one or more the highest deduction of text chunk.Deduction is in communication meeting The information that can show on the computing device during words, for transmitting the theme of content during communication session to user Time, it is provided that other relevant with the theme of content or more details.Deduction can also be at the communication process with the equipment of calculating In be sent to remote equipment and show on a remote device.When needing other process for other word or text chunk, block 105 can be after block 101.
During determining deduction, the use of network search engines allows other Advanced Options.Such as, draw based on search The virtual assistance system holding up deduction is configured such that from web search results the most specific in deleting communication session Term.Thus, if there being people film star being discussed and is mentioning the name of this film star during communication session, then based on The assistance system that search engine is inferred can get rid of this specific name, it will be seen that there are many common ground with specific people But the actual more information relevant with he incoherent film stars.Political or commercial, this can allow people in real time Counter-example or other option are provided, thus form negotiation.
Web search results can also be set to only with nearest result as target, thus collect only with nearest news Relevant material.This topic known before people can be helped to follow the trail of he/her a period of time, or people can be helped to catch up with communication The dialogue of session.
Fig. 2 A shows based on search engine deduction according to arrange according at least some embodiment described herein Virtual assistance, for determining the instantiation procedure of deduction as the input of calculating equipment from the word received during communication session The flow chart of 200.In shown example, can be described as processing the mistake of step, feature operation, event and/or behavior etc. Journey 200 and other process as described herein, various functional block or behavior can be passed through hardware, software and/or firmware and perform.This Skilled person according to the disclosure it will be appreciated that the various of the functional block shown in Fig. 2 can be put into practice in various embodiments Modification.Such as, although process as shown in Figure 2 200 includes one specific order of piece or behavior, but these blocks or behavior institute Theme required for protection need not be confined to any particular order by the order presented.Similarly, without departing from claimed Theme scope in the case of, it is possible to use the behavior between two parties not shown in Fig. 2 and/or the adjunctive behavior not shown in Fig. 2, And/or some behaviors shown in Fig. 2 can be removed.Process 200 can include operating 103,201,202,203,205,207,209 And/or one or more in 105, wherein can start to process at block 201.
At block 201, " receiving word from speech-to-text engine ", receive 201 words from speech-to-text engine.Block 202 is permissible After block 201.At block 202, " the rare degree of word is on predetermined threshold?", when receiving 200 and arriving each word, the most really Whether the rare degree of fixed 202 words is on predetermined threshold.Word rarity scoring can be initially applied to word.High word rarity is commented Dividing and will imply that, this word is the most rare, and low word rarity scoring will imply that, this word is fairly common.For example, it is possible to storage And access the most common word primary list (e.g., " if ", " being somebody's turn to do " and "Yes").If this word occurs in the list of the most common word On, the most such as this word can be ignored completely, or give the rare degree scoring of this word 0 point.If this word does not appears in the most normal Seeing in the list of word, then can access stored word rarity list, this list has being correlated with for word each in list Predetermined word rarity is marked, to determine whether this word occurs in word rarity list and the relevant word rarity of this word is commented Point.
If this word does not appears in word rarity list, then this word can be used as the search art in web search Language performs to use WWW (network) search of network search engines (such as, Google) automatically.The hit returned from this search Quantity can serve as the basis of the word rarity scoring of this word.The hit returned is the fewest, and the word rarity for this word is marked more High.Such as, make word " counselor " as the search term in web search, 68,400,000 hits can be produced, make 9,000,000 hits can be returned with " Trinity university ".But, word " instrument " can return and hit without several, thus returns remote Word rarity scoring less than " counselor " or " Trinity university ".It is then possible to word and the scoring of relevant word rarity are added Add to word rarity list to use in the future.
If the rare degree of word not on predetermined threshold (such as, do not have higher than scheduled volume word rarity mark), Then this process is forwarded to next word received, and for this next one word repeat the above steps.Additionally, specific threshold can be User can carry out configuring by the one or more user interfaces on calculating equipment.If the rare degree of word predetermined threshold it Upper (such as, have the word rarity higher than scheduled volume), then be labeled as rare by this word, it is possible to any adjacent adjective Or adverbial word combines.
Block 203 can be after block 202.At block 203, and " by word add to inferring nomenclature ", then by produced word/ Phrase adds 203 to inferring nomenclature.Thus, when receiving each word, one or more phase can be stored in a buffer Adjacent word, to combine them in this manner.By rare word and adjacent adjective and adverbial word composition group, even if individually Adjective and adverbial word are common.Such as, this allows the difference between people's capture " bad business " and " good business ", and this can cause Visibly different deduction, even if word " bad " and " good " self is the most common to such an extent as to can be dropped.
Block 205 can be after block 203.At block 205, " determine deduction quantity ", based on the word used in web search/ Phrase, automatically determines the quantity of 205 most common deductions.For example, it is possible to by making word/phrase as searching in web search Suo Shuyu, and correspondingly analyze Search Results, determine deduction quantity.For example, it is possible to automatically read search from Search Results The topic of result, and it is used as the deduction for this word/phrase.Infer that quantity can be the different topics found in Search Results Quantity (or quantity of the different topics in front n Search Results).Most common deduction in Search Results is and search knot In fruit, other infers that comparing those the most often occurred in Search Results infers.The analysis from web search results can be used for In obtain infer other automatic technologies various and process be embodied as those above process alternate item or with those above mistake Journey combines.
Furthermore, it is possible to compared with the word/phrase being search for term by the relevant deduction that Search Results is exported Determine deduction intensity.Such as, if the web search for word/phrase A returns Search Results output B, then can automatically search Rope B, to determine that the search hit of many vast scales also includes A.If a high proportion of search hit includes A, it is determined that infer intensity The search hit being better than minor proportion includes the situation of A.
Block 207 can be after block 205.At block 207, " infer that quantity is under threshold value?", determine 207 for word/phrase Deduction quantity and/or web search results hit whether under specific threshold.Can based on deduction quantity determined above, Word/phrase is made to make overall relevance of searches as the web search results quantity of search term or its any combination Determine.Such as, if the different excessive number inferred found in Search Results, then this indicates any specific deduction would is that Relevant or useful probability is relatively low.If additionally, the excessive number of the web search results of word/phrase (that is, hit), then this refers to Show that word/phrase is the most common to such an extent as to its own cannot be used to determine useful deduction.
Relation between different quantity and the quantity that Search Results hits inferred found or different pushing away of being found The disconnected ratio accounting for Search Results hit is also used as determining whether to meet the factor of specific threshold.Additionally, specific threshold is permissible It is that user can carry out configuring by the one or more user interfaces on calculating equipment.
Block 105 can be after block 207.At block 105, " showing the highest deduction ", if for the deduction of this word/phrase And/or the quantity of web search results is under specific threshold, the most on the computing device according to real-time constraint or close to the most about Bundle, display 105 is for the highest deduction (seeing Fig. 3 as example) of this word/phrase.Deduction can also provide relevant with deduction The link of additional information, and can include and infer relevant advertisement etc..Then, this process repeats, to receive 200 communications The step of the next word in session starts (such as, from the beginning of speech-to-text engine)
Deduction can also be automatically fed to speech-to-text or speech-to-text identification system, to improve the solution of this system Translate performance.Such as, English phrase " a narrow flight " and " an arrow flight " is even for being in quiet room In people for be also undistinguishable, but if based on search engine infer virtual assistance system make deduction, speak Person has just talked about Errol Flynn during communication session, then the words is it is more likely that about " arrow ".If based on The virtual assistance system that search engine is inferred makes deduction, and speaker has just talked about Dennis Lau during communication session (architect), then the words is likely about stair, and this can be based on term relevance of searches each other and determines 's.
Block 209 can be after block 207.At block 209, " portmanteau word/phrase ", if for this word/phrase deduction and/ Or the quantity of web search results hit is on specific threshold, then by this word/phrase and or many inferred in nomenclature Other adjacent word individual/phrase combination 209, to produce bigger phrase.It is previously determined adjacent word/phrase the dilutest Have to add in this deduction nomenclature, but perhaps itself be unsatisfactory for making any relevant threshold value inferring display.If additionally, Adjacent word not yet adds deduction nomenclature (such as, adjacent word is on the right side of current phrase) to, then can store from being used for The buffer of adjacent word reads this adjacent word, or this process can wait until and be added to by other deduction term Infer nomenclature.
Such as, if for the quantity of the deduction of word " instrument " and/or Search Results hit on specific threshold, then will This word with infer in nomenclature that the adjacent word " power " in word " instrument " left side is combined 209, to produce longer phrase " merit Rate instrument ".
Then, this process branch, and automatically repeat, to use new portmanteau word/phrase to determine 205 in web search The quantity of most common deduction starts, the most also to receive next word (such as, the formation speech-to-text in 201 communication sessions Engine) start.In above example, phrase " power tool " will act as the search term in web search, meeting on this surface The different of the quantity more less than the situation of individually " instrument " are caused to be inferred and the hit of total Search Results.Along with process continues, Word/the phrase used in web search can be increased, until inferring and/or net by all adding adjacent word along both direction The quantity of network Search Results hit is under the specific threshold that display is inferred, or there is not additional word and to be combined.Optional Ground, can continue to increase use in web search by only adding adjacent word to the left side of current word/phrase or right side Word/phrase.Furthermore, it is possible to the word to be appended such as have during communication session to become phrase before available (such as, being uttered) The restriction of which kind of degree can be increased to.
Fig. 2 B shows based on search engine deduction according to arrange according at least some embodiment described herein Virtual assistance, determines the example alternative procedure of deduction as the input of calculating equipment for the word received during communication session The flow chart of 210.
For example, it is possible to utilize the process 210 of Fig. 2 B to receive from by a text chunk single treatment stream text To word in determine deduction, it is different from the process shown in Fig. 2 A, wherein a word in single treatment stream text.Text chunk Size can change, and can be selected by user.At block 211, " receiving text chunk from speech-to-text engine ", Ke Yicong Speech-to-text engine receives text chunk.Block 213 can be after block 211.At block 213, " selecting rare word ", once from voice 211 text chunks are received, then as set forth above, it is possible to by the rare word of text chunk with any to text engine or other communications applications Applicable adjacent adjective or adverbial word combine (generation phrase).As set forth above, it is possible to reference to Fig. 2 A, by word rarity is commented Distribute to the word in text chunk and only select that there are those words marked higher than the word rarity of predetermined threshold, selecting 213 Rare word.
Block 215 can be after block 213.At block 215, " determining topic relevance ", for selected word/phrase and text chunk Each combination of interior adjacent word/phrase, by using this combination to determine 215 combinations as the search term in web search Topic relevance.Quantity based on the hit received in web search, it may be determined that the topic of word/phrase in 215 combinations Relevance.Generally, the hit quantity produced from web search is the lowest, and in combination, the topic relevance of word/phrase is the lowest.
Block 217 can be after block 215.At block 217, " topic relevance is under predetermined threshold?", can based on for Hit quantity in the Search Results of combination determines that the topic relevance of 217 any combination is whether under predetermined threshold.
Block 219 can be after block 217.At block 219, " producing bigger selected phrase ", if based in Search Results , there is not the combination under predetermined threshold of the topic relevance in the quantity of hit, then by by one or more adjacent words with Selected word/phrase combination, produces 219 bigger selected phrases.This process can start with word/phrase selected by first, by will It produces bigger word/phrase for further combination with adjacent selected word/phrase combination.Then, with for selected Each combination of word/phrase word/phrase adjacent with in text chunk determines that the step of the topic relevance of 215 these combinations is opened Begin, use bigger word/phrase combination to repeat this process.
Block 221 can be after block 217.At block 221, " adding word/phrase combination as inferring term ", if based on searching In hitch fruit, there is topic relevance any combination under predetermined threshold, then can add 221 and can apply in the quantity of hit Word/phrase (that is, its topic relevance those word/phrases under predetermined threshold) is as the deduction art inferred in nomenclature Language.Following present example and infer nomenclature.
As shown in above-mentioned deduction nomenclature, list each word/phrase at the top of table and side.In can being arranged by lookup A word/phrase in the combination of word/phrase, and through this table until the row of another word/phrase arrived in word/phrase combination, The topic relevance of the combination of word/phrase can be found in table.Then, the position of obtained table is the network according to combination Search is hit an instruction of the topic relevance of combination for quantity.Hit for the suitable small number of particular combination It is also used as this combination and will have generally indicating of higher deduction intensity.
Block 223 can be after block 221.At block 223, " find the most incoherent deduction of topic ", it appeared that 223 for Infer the deduction that the topic of each combination inferring term in nomenclature is least correlated with.System can be selected to be to select by user The most incoherent deduction of single topic or the most incoherent deduction of multiple topic.Can will be used for dividing from web search results Analysis obtains other automatic technologies various of inferring and process is embodied as alternate item or and the those above of those above process Process combines.
Furthermore, it is possible to by checking, relative to the word/phrase being search for term, the relevant deduction that Search Results exports Determine deduction intensity.Such as, if the web search for word/phrase A returns Search Results output B, then can automatically search Rope B, to determine that the search hit of many vast scales also includes A.If a high proportion of search hit includes A, it is determined that infer intensity The search hit being better than minor proportion includes the situation of A.
Block 225 can be after block 223.At block 225, " infer that intensity is on threshold value?", determine 225 for each group Whether the intensity of the deduction that the topic closed least is correlated with is on specific threshold, and/or the web search of combination returns specific The hit of the quantity under threshold value.Determine that the intensity web search hit whether on specific threshold or for combination is No under specific threshold or the two depends on that the desired qualities of deduction finally shown and accuracy infer number relative to expectation Amount.If intensity to be determined whether on specific threshold and for combination web search hit whether specific threshold it Under, then can improve quality and the accuracy of the deduction of final display, but the quantity of shown deduction can be reduced.Threshold level Also this is had an impact, because the higher thresholds for intensity and the lower threshold for hit quantity also result in final display The quality inferred and the raising of accuracy, but the quantity of shown deduction can be reduced.Can select to determine by force by user Degree whether on specific threshold and/or for the web search hit of combination whether under specific threshold and threshold value water Flat own.
If deduction intensity is not on specific threshold, and/or infer that the web search of term combination does not return specific The hit of the quantity under threshold value, then block 219 can be after block 225.Said process with by by text chunk or Multiple adjacent words and selected word/phrase combination produce the step of 219 bigger selected phrases and start to repeat.
Block 227 can be after block 225.At block 227, " showing the highest deduction ", if infer intensity specific threshold it Upper and/or deduction term combination web search returns the hit of the quantity under specific threshold, the most real-time Or show the 227 the highest deduction combined for word/phrase or deductions close to real-time constraint.Can infer based on as above Intensity and common point determine and select the highest deduction.The quantity of the highest deduction to be shown or the most only show the highest deduction Can also is that user's configurable features of the virtual assistance system inferred based on search engine.
After block 227, this process is to open from speech-to-text engine or other communications applications reception 211 next text chunk Starting weight is multiple.However, if it is desired to, can one or more in inferring in nomenclature the text chunk that leave previously processed mistake Word/phrase, in order to by considering that the word/phrase of the text chunk from previously processed mistake increases the width of analyzed text. Such as, it is retained in word/the phrase inferred in nomenclature among the text chunk of previously processed mistake can be and use this word/phrase As those the rarest word/phrases that the search web search results quantity that returned of term is relevant.Furthermore, it is possible to exist For identical deduction display number of times restriction, the display of this identical deduction can cause combine from the word in new text chunk/ Phrase and using is previously added to infer the word/phrase of nomenclature.
Fig. 3 is according at least some embodiment as described herein, based on real-time reception in mobile device 301 logical The diagram of the virtual assistance that letter, basis are inferred based on search engine, display example deduction 317 example mobile device 301.Institute The mobile device 301 illustrated has mobile device shell 303, display screen 305, user's load button 323 and audio frequency output 325.Mobile device can have internal hardware, computer-readable medium and be stored thereon to perform various radio communication And run computer software programs code be suitable for application, data and computer-readable instruction, with for based on search engine The virtual assistance inferred performs method as described herein and process.With further reference to Fig. 4 to 6 describe this for for The virtual assistance inferred based on search engine performs method as described herein and the hardware of process and computer-readable medium The example calculating equipment being suitable for other.
The display screen 305 of mobile device 301 shows and performs for the example communication session in mobile device 301 The example of virtual assistance inferred based on search engine.Example communication session shows by providing audio frequency to equipment 301 It is sent to video or the Video chat of still image of the long-distance user 309 of mobile device 301 with vision signal and/or data Session.Audio signal and/or data that equipment 301 receives can be by residing in (example in mobile device 301 as mentioned above As, in the storage device of mobile device, this storage device can be to include but not limited to volatile memory (such as RAM), non- Volatile memory (such as ROM, flash memory etc.) or its any combination any type) speech-to-text engine and this locality is converted to Text or can previously have been carried out conversion, be then sent collectively to mobile device 301 as text together with audio frequency. Show the corresponding text such as flowed in the documentwindow 307 on display screen 305 when long-distance user 309 talks. Identify during the virtual assistance inferred based on search engine as mentioned above each deduction term 311,313,315 or its During combination, can they are highlighted, insertion of brackets etc., the deduction term being designated as in documentwindow 307 or combination.Can also be at text Near applied deduction term 311,313,315 in window 307, display uses corresponding deduction term to return as search term The correlated measure of the web search hit returned.Such as, infer that term " counselor " 311 returns 68,400,000 networks and searches Rope hits, and infers that term " Trinity university " 313 is returned only to 9,000,000 hits.
Show when communication session occurs in mobile device 301 according to real-time or constrain in display screen close to real-time Show on curtain 305 that example infers 317.Under documentwindow 307 on display screen 305, " discussion is appraised through discussion in the specific deduction of display Member John Cornyn, TX " 317 it is to be performed for the present communications session illustrated in mobile device 301 by described above The virtual assistance process inferred based on search engine in real time or close to the deduction term " counselor " identified in real time and process 311, " Trinity university " 313 and the result of " judge " 315.Can also show together with inferring 317 or near deduction 317 Additional information, as to about this deduction in further detail or the link 319 of relevant information and the relative intensity of this deduction 321.Together with deduction 317 or near deduction 317, the out of Memory of display can include but not limited to and infer that 317 is relevant Advertisement, storage or process further infer 317 option and with infer 317 other relevant statistics.
Fig. 4 is the diagram of example networked computing environment 400, wherein can realize many Computerized procedures to perform basis The virtual assistance inferred based on search engine of at least some example described herein.Such as, communication session described above can To occur between each object in the networked computing environment shown in Fig. 4, the one or more objects in Fig. 4 can utilize or Realize the virtual assistance inferred based on search engine.As another example, distributed or parallel calculating can be this networking A part for environment, wherein the various clients on Fig. 4 network use and/or realize virtual for infer based on search engine The system and method assisted.It will be understood by those skilled in the art that network may be used for any computer or other client Or server apparatus with work alone or other computer in a distributed computing environment or other client or server set Standby connection.In this respect, it is believed that there is any number of process, memory or memory cell and the Arbitrary Digit simultaneously occurred The application of amount and any computer system of process or environment are suitable to combine the system and method provided and use.
Distributed Calculation may be used for by swapping shared calculating money between various calculating equipment and/or system Source and/or service.These resources and/or service can include exchange information, use cache memory and/or use for The disk storage of distributed document storage.Distributed Calculation can utilize network connectivity, it is allowed to client regulates being total to of they With power so that whole enterprise is benefited.In this respect, various equipment can have can contain as described herein based on search The application of virtual assistance process, object or the resource that engine is inferred.
The networked computing environment 400 of Fig. 4 can include various equipment, as one or more calculating equipment 271,600,276, 301, one or more objects 273,274 and 275 and/or one or more database 278.These equipment 271,600,273, 274, each in 275,276,301 and 278 can include or utilize the storage of program, method, data, FPGA etc.. Calculating equipment 301 can be the mobile device 301 of Fig. 3.But, equipment 271,600,273,274,275,276,301 and/or 278 can cross over identical or different equipment (as personal digital assistant (PDA), mobile device, audio/video devices, MP3 play Device, personal computer etc.) each several part.Each equipment 271,600,273,274,275,276,301 and 278 may be adapted to pass through Communication network 270 communicates with another equipment 271,600,273,274,275,276,301 and 278.In this respect, arbitrary equipment can With responsible database 278 or the maintenance of other memory element and renewal.
Communication network 270 self can include other computational entity, and this computational entity is configured to provide to the system of Fig. 4 Service, thus multiple interference networks can be represented.According to the aspect of some embodiments, each equipment 271,600,273,274, 275,276,301 and/or 278 can comprise discrete function program module, it is possible to use API (API) or other is right As, software, firmware and/or hardware, be suitable to ask in miscellaneous equipment 271,600,273,274,275,276,301 and/or 278 One or more services.
It is further appreciated that object (such as 275) can live with on another calculating equipment 276.Thus, although described thing Connection equipment can be shown as computer by reason environment, but this explanation is only example, and physical environment can be described alternatively For including various digital device, such as PDA, TV, MP3 player etc. and software object, such as interface, com object etc..
Existence can support various systems, assembly and the network configuration of DCE.Such as, equipment is calculated permissible Connected together by wired or wireless system, LAN or wide-area distribution type network.At present, many networks and internet coupling Closing, internet is the foundation structure calculated for wide-area distribution type, and includes many different networks.Any this network (nothing Whether opinion couples with internet) all can use in conjunction with the system and method provided.
Network infrastructure can realize the main frame of network topology, such as client/server, equity or mixed structure." visitor Family end " it is class or the member of group using service that is the most incoherent another kind of or that organize.In the calculation, client be ask by The process of the service that another program provides, the most roughly, instruction or task-set.Client process utilizes the service asked, and Any operational detail relevant with other program or service itself " need not be known ".In client terminal/server structure, especially In networked system, client typically accesses the calculating of the shared Internet resources provided by another computer (such as, server) Machine.In the example of fig. 4, according to environment, arbitrary equipment 271,600,273,274,275,276,301 and/or 278 can be recognized For be client, server or the two.
Will typically, but not necessarily, server is the long-range meter that can be accessed by remotely-or locally network (e.g., internet) Calculation machine system.Client process can be active in first computer system, and server process can be at second computer System is active, by communication media with communicate with one another, thus distributed function is provided and allows multiple client to utilize The information gathering capability of server.Arbitrarily software object can be distributed on multiple calculating equipment or object.
Client and server may be adapted to utilize the function provided by protocol layer to communicate with one another.Such as, hypertext passes Transmission protocol (HTTP) is the puppy parc that can use in conjunction with WWW (WWW) or " network ".Such as Internet Protocol (IP) Other of the computer network address of location etc or such as resource locator (URL) etc is quoted and be may be used for identifying each other Server or client computer.The network address can be referred to as URL address.Communication, example can be provided by communication media As, for large-capacity communication, client and server can connect coupled to each other via TCP/IP.
According to can according in Fig. 4 provide general framework build different computing environments and can be such as Fig. 4's Other variant occurred in calculating in network environment, system and method provided here is not construed in any way It is limited to specific computation structure.But, it should on width and scope, explain embodiment according to claims.
Fig. 5 shows that at least some embodiment according to the disclosure is arranged, for the void inferred based on search engine Intend the schematic diagram of the computer program assisted.Computer program 500 can include for perform above-mentioned and Fig. 1, One or more executable instruction set 502 of the method shown in Fig. 2 A and Fig. 2 B.Computer program 500 can be at list Individual bearing medium 504 or another like communication media 506 transmit.Computer program 500 can also record at computer In computer-readable recording medium 508 or another like computer-readable recording medium 510.
Fig. 6 be that at least some embodiment according to the disclosure is arranged, can realize thereon inferring based on search engine The block diagram of Example Computing Device of virtual assistance.In the most basic configuration, calculating equipment 600 typically comprises one or many Individual host-processor 604 and system storage 606.Memory bus 608 may be used in host-processor 604 and system Communicate between memory 606.
According to expectation configuration, host-processor 604 can be any type, includes but not limited to microprocessor (μ P), micro- Controller (μ C), digital signal processor (DSP) or its any combination.Processor 604 can comprise more than the high speed of one-level and delays Deposit, such as on-chip cache device 610 and secondary cache 612, processor cores 614 and register 616.At example Reason device kernel 614 can include ALU (ALU), floating point unit (FPU), digital signal processing core (DSP core), Or its any combination.Example storage controller 618 can also be used together with processor 604, or in certain embodiments, deposits Storage controller 618 can be the interior section of processor 604.
According to expectation configuration, system storage 606 can be any type, includes but not limited to that volatile memory is (e.g., RAM), nonvolatile memory (e.g., ROM, flash memory etc.) or its any combination.System storage 606 can include operating system 620, one or more application 622 and routine data 624.In some embodiments, operating system 620 can have tune Degree device 626, and operation one or more application 622 are set, to perform function as described herein, including at least with figure 1, the function of the process prescription shown in Fig. 2 A and Fig. 2 B.Additionally, application 622 can be configured so that together with routine data 624 Operate in operating system 620.Routine data 624 can include task relevant information, such as but not limited to, it is used for performing with operation The task data that the instruction of virtual assistance based on search engine deduction is correlated with.In figure 6, by those groups in inner dotted line Part illustrates basic configuration 602 as described herein.
Calculating equipment 600 can have supplementary features or function and additional interface, is used for being easy to basic configuration 602 He The arbitrarily communication between equipment needed thereby and interface.Such as, bus/interface controller 630 may be used for being easy to via memory interface Bus 634, communication between basic configuration 602 and one or more data storage device 632.Data storage device 632 Can be removable storage device 636, non-removable storage device 638 or a combination thereof.Removable storage device and non-removable The example of storage device include the disk unit such as disk drive and hard drive (HDD) etc, as compact-disc (CD) drive or Digital universal disc (DVD) drives, the disc drives of solid-state driving (SSD) and band driving etc.Example computer storage media Any side for storing information (such as computer-readable instruction, data structure, program module or other data) can be included in The volatibility realized in method or technology and non-volatile, removable and nonremovable medium.
System storage 606, removable storage device 636 and non-removable storage device 638 are computer-readable storage mediums Example.Computer-readable storage medium includes but not limited to RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, number Word universal disc (DVD) or other optical memory, tape, magnetic disc store or other magnetic storage apparatus or may be used for storage expectation Information other medium any that can be accessed by calculating equipment 600.Any this computer-readable storage medium can be calculating equipment The part of 600.
Calculating equipment 600 can also include interface bus 640, for be easy to via bus/interface controller 630 from respectively Plant interface equipment (such as, output equipment 642, Peripheral Interface 644 and communication equipment 646) communicating to basic configuration 602.Example Output equipment 642 includes GPU 648 and audio treatment unit 650, and they can be configured to via one or many Individual A/V port 652 and the various external device communications of such as display or loudspeaker etc.Example Peripheral Interface 644 includes string Line interface controller or parallel interface controller, they can be configured to via one or more I/O ports 658 with the most defeated Enter external equipment or other peripheral hardwares of equipment (such as, keyboard, mouse, pen, voice-input device, touch input device etc.) etc (such as, printer, scanner etc.) communicates.Example communication device 646 includes network controller, can be configured so that be easy to via One or more COM1s, communicate with other calculating equipment 622 one or more on network communication link.Implement at some In mode, calculating equipment 600 includes polycaryon processor 664, can be communicated with host-processor 604 by interface bus 640.
Network communication link can be an example of communication media.Communication media typical case can be referred to by computer-readable Make, data structure, program module or other data (e.g., carrier wave or other transmission mechanism) in modulated data signal concrete Change, it is possible to include random information delivery media." modulated data signal " can be such signal, and one or more is special Levy and be set or changed to encode information onto in the signal.As example without limitation, communication media can include Line medium, as cable network or straight line connect and wireless medium, such as sound, radio frequency (RF), microwave, infrared (IR) and other nothing Line medium.Term computer-readable medium used herein above can include storage medium and communication media.
Calculating equipment 600 can be implemented as small portable (or mobile) electronic equipment (e.g., mobile phone, personal digital assistant (PDA), personal media player equipment, wireless network facilities for observation, personal mobile phone equipment, special equipment or include any of the above The mixing apparatus of function) a part.Calculating equipment 600 can also be embodied as including laptop computer and non-calculating on knee The personal computer of machine configuration.
There are some little difference between the hardware and software implementation of system schema;The use of hardware or software is general (but and not always, because the selection under specific circumstances between hardware and software may become critically important) is a kind of to embody cost And the design alternative of balance between efficiency.Can implement to be retouched here with various means (such as, hardware, software and/or firmware) The technique stated and/or system and/or other technologies, and preferably technique will along with described technique and/or system and/or other Environment that technology is applied and change.Such as, if realization side determines that speed and accuracy are most important, then realization side is permissible Select predominantly hardware and/or the means of firmware configuration;If flexibility is most important, then realization side can select mainly The embodiment of software;Or, the most alternatively, it is achieved side can select specific group of hardware, software and/or firmware Close.
Above detailed description, by user's block diagram, flow chart and/or example, has elaborated equipment and/or technique Numerous embodiments.In the case of this block diagram, flow chart and/or example comprise one or more function and/or operation, It will be understood by those skilled in the art that each function in this block diagram, flow chart or example and/or operation can be by various Hardware, software, firmware or substantially their any combination individually and/or jointly to realize.In one embodiment, the disclosure If the stem portion of described theme can pass through special IC (ASIC), field programmable gate array (FPGA), data signal Processor (DSP) or other integrated forms realize.But, those skilled in the art will appreciate that enforcement disclosed herein Some aspects of example can realize in integrated circuits on the whole or partly equally, it is achieved for calculating at one or more The one or more computer programs run on machine are (such as, it is achieved for one run in one or more computer system Or multiple program), it is achieved for one or more programs of running on the one or more processors (such as, it is achieved at one Or the one or more programs run on multi-microprocessor), it is achieved for firmware, or substantially it is embodied as appointing of aforesaid way Meaning combination, and those skilled in the art are according to the disclosure, will be provided with designing circuit and/or write software and/or firmware code Ability.Additionally, the mechanism that it would be recognized by those skilled in the art that theme described in the disclosure can be as the program of various ways Product is distributed, and is used for performing the particular type of the signal bearing medium of distribution regardless of reality, described in the disclosure The exemplary embodiment of theme is the most applicable.The example of signal bearing medium includes but not limited to: recordable-type media, as floppy disk, Hard disk drive, compact-disc (CD), digital video disc (DVD), digital magnetic tape, computer storage, flash memory etc.;And mode transmission Medium, such as numeral and/or analogue communication medium (such as, optical fiber cable, waveguide, wired communications links, wireless communication link etc.).
Those skilled in the art will appreciate that above-detailed equipment and/or technique, hereafter use engineering practice Described equipment and/or technique being integrated in data handling system is the conventional means of this area.That is, it is described here Can being integrated in data handling system by the test of fair amount at least partially of equipment and/or technique.Ability Field technique personnel it will be recognized that typical data handling system generally comprise in the following one or more: system unit Shell;Video display apparatus;Memory, such as volatibility and nonvolatile memory;Processor, such as microprocessor and data signal Processor;Computational entity, such as operating system, driver, graphical user interface and application program;One or more alternately Equipment, such as touch pad or screen;And/or control system, including feedback loop and control motor (such as, be used for sensing position and/ Or the feedback of speed;For moving and/or adjust the control circuit machine of component and/or quantity).Typical data handling system can To utilize the commercial parts (as data calculate/communicate and/or parts conventional in network calculations/communication system) of any appropriate to give To realize.
Theme described in the disclosure sometimes illustrate different parts be included in different miscellaneous parts in or different parts with Different miscellaneous parts is connected.Should be appreciated that the framework so described is example, in fact can realize many and be capable of Other frameworks of identical function.Conceptually, effectively " associate " any of the parts in order to realize identical function and arrange, thus Realize required function.Therefore, combination here realize any two parts of concrete function can be considered to be " associated with " each other thus Realize required function, regardless of whether framework or intermediate member are how.Equally, the parts that any two so associates can also be regarded as " it is operably connected " each other or " being operatively coupled " is to realize required function, and any two parts that can so associate Can also be considered " can be operatively coupled " to realize required function each other.The concrete example that can be operatively coupled include but It is not limited to physically can connect and/or the most mutual parts, and/or the parts of wireless interaction and/or wireless interactive, And/or logic is mutual and/or can the mutual parts of logic.
To the most any about most and/or the use of singular references, those skilled in the art can be from many number forms Formula is converted to singulative, and/or is converted to most form from singulative, with applicable specific environment and application.For understanding See, clearly state that at this singulative/majority form is interchangeable.
It will be appreciated by those skilled in the art that it is said that in general, the term that used, particularly (example in claims As, in the main part of claims) term that uses, usually it is interpreted as open to the outside world term (such as, term " include " should be interpreted that " including but not limited to ", term " have " should be interpreted that " at least having " etc.).Those skilled in the art are also Should be understood that if being intended in introduced claim indicate specific number, the most this intention will in this claim clearly Point out, and do not have this clearly indicate in the case of, the most there is not this intention.Such as, for helping to understand, appended right is wanted Ask and may have been used the feature guiding phrase " at least one " and " one or more " to introduce in claim.But, this The use of phrase is not necessarily to be construed as imply the claim feature introduced by indefinite article "a" or "an" will comprise this Any specific claim of feature is limited to only comprise the invention an of this feature, even if this claim both includes guiding Phrase " one or more " or " at least one " include again indefinite article such as "a" or "an" (such as, " " and/or " " Should typically be interpreted to mean " at least one " or " one or more ");Introduce in claim using definite article Feature time, same.Even if it addition, having explicitly pointed out the specific number of introduced claim feature, art technology Personnel it should be understood that this enumerating should be typically interpreted to mean that at least institute's column number (such as, does not exist other modifiers Phrase " two features " typically means at least two this feature, or two or more this feature).It addition, it is similar using In the case of " at least one in A, B and C etc. " such statement, in general generally should manage according to those skilled in the art The implication solving this statement makes an explanation and (such as, " has the system of at least one in A, B and C " and should include but not limited to individually There is A, individually there is B, individually there is C, there is A and B, there is A and C, there is B and C and/or there is the system etc. of A, B, C). In the case of use is similar to " at least one in A, B or C etc. " such statement, in general should be according to art technology Personnel be generally understood that the implication of this statement make an explanation (such as, " there is the system of at least one in A, B or C " and should include but It is not limited to individually to have A, individually there is B, individually there is C, there is A and B, there is A and C, there is B and C and/or there is A, B, C System etc.).It should also be understood by those skilled in the art that the adversative conjunction the most arbitrarily representing two or more optional terms And/or phrase, either in specification, claims or accompanying drawing, shall be construed as giving and include these terms One of, these terms either one or the possibility of two terms.Such as, phrase " A or B " be understood to include " A " or " B " or the possibility of " A and B ".
Although being disclosed that multiple scheme and embodiment, but it will be appreciated by those skilled in the art that other schemes And embodiment.Multiple scheme disclosed herein and embodiment are for illustrative purposes rather than restrictive, the disclosure True scope and spirit characterized by claims.

Claims (20)

1., for automatically being shown the method inferring term by calculating equipment, the method includes:
Receive streaming audio data or stream video data;
When receiving described streaming audio data or stream video data, streaming audio data that substantially real-time broadcasting receives or stream Video data;
When the streaming audio data received by playing or stream video data, show corresponding to received streaming audio data or The stream text of the voice in stream video data;
Identify the deduction term in described stream text;
Indicate the deduction term in the described stream text identified intuitively, wherein, indicate the described stream literary composition identified intuitively Infer that term includes highlighted to the deduction term identified in the documentwindow comprising described stream text or adds and include in Ben Number;And
Together with the deduction term identified, show the web search message relevant to the deduction term identified.
Method the most according to claim 1, wherein, described calculating equipment includes mobile device.
Method the most according to claim 1, wherein, receives described streaming audio data or stream video counts in a communication session According to.
Method the most according to claim 1, wherein, described streaming audio data or stream video data are included in Video chat meeting The video data received in words.
Method the most according to claim 1, the method also includes: uses and is configured at received voice data or regards Frequency produces described stream text according to the speech-to-text engine of upper operation.
Method the most according to claim 1, the method also includes: together with described streaming audio data or stream video data Ground receives described stream text.
Method the most according to claim 1, wherein, shows that described stream text is included in display stream text in documentwindow.
Method the most according to claim 1, wherein, identifies that the deduction term in described stream text includes identifying described stream literary composition Rare word in Ben.
Method the most according to claim 1, wherein, shows the web search message relevant to the deduction term identified Including display by the quantity of the web search hit produced about the web search inferring term identified.
Method the most according to claim 1, the method also includes: display is inferred, described deduction uses that is identified to push away Disconnected term identifies.
11. 1 kinds calculate equipment, and this calculating equipment includes:
Processor;
Memory;And
The virtual assistance application inferred based on search engine, the application of described virtual assistance is configured that
Receive streaming audio data or stream video data;
When receiving described streaming audio data or stream video data, streaming audio data that substantially real-time broadcasting receives or stream Video data;
When the streaming audio data received by playing or stream video data, show corresponding to received streaming audio data or The stream text of the voice in stream video data;
Identify the deduction term in described stream text;
Indicate the deduction term in the described stream text identified intuitively, wherein, indicate the described stream literary composition identified intuitively Infer that term includes highlighted to the deduction term identified in the documentwindow comprising described stream text or adds and include in Ben Number;And
Together with the deduction term identified, show the web search message relevant to the deduction term identified.
12. calculating equipment according to claim 11, wherein, described calculating equipment includes mobile device.
13. calculating equipment according to claim 11, wherein, the described virtual assistance application inferred based on search engine is joined It is set to receive described streaming audio data or stream video data in a communication session.
14. calculating equipment according to claim 11, wherein, the described virtual assistance application inferred based on search engine is joined It is set in video chat session, receive described streaming audio data or stream video data.
15. calculating equipment according to claim 11, wherein, the described virtual assistance application inferred based on search engine is joined It is set to, uses the speech-to-text engine being configured on received voice data or video data operation to produce described Stream text.
16. calculating equipment according to claim 11, wherein, the described virtual assistance application inferred based on search engine is joined It is set to receive described stream text together with described streaming audio data or stream video data.
17. calculating equipment according to claim 11, wherein, the described virtual assistance application inferred based on search engine is joined It is set to by showing that in documentwindow stream text shows described stream text.
18. calculating equipment according to claim 11, wherein, the described virtual assistance application inferred based on search engine is joined It is set to the deduction term identifying in described stream text by the rare word in the described stream text of identification.
19. calculating equipment according to claim 11, wherein, the described virtual assistance application inferred based on search engine is joined It is set to, is shown by the quantity of the web search hit produced about the web search inferring term identified by display The web search message relevant to the deduction term identified.
20. calculating equipment according to claim 11, wherein, the described virtual assistance application inferred based on search engine is joined Being set to display infer, the described deduction term inferring that use is identified identifies.
CN201610082319.2A 2010-03-16 2011-03-07 Method for automatically displaying inferences and computing device Pending CN105760464A (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US12/724,660 US8214344B2 (en) 2010-03-16 2010-03-16 Search engine inference based virtual assistance
US12/724,660 2010-03-16

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
CN201180004870.0A Division CN103168298B (en) 2010-03-16 2011-03-07 Based on the virtual assistance that search engine is inferred

Publications (1)

Publication Number Publication Date
CN105760464A true CN105760464A (en) 2016-07-13

Family

ID=44648030

Family Applications (2)

Application Number Title Priority Date Filing Date
CN201180004870.0A Expired - Fee Related CN103168298B (en) 2010-03-16 2011-03-07 Based on the virtual assistance that search engine is inferred
CN201610082319.2A Pending CN105760464A (en) 2010-03-16 2011-03-07 Method for automatically displaying inferences and computing device

Family Applications Before (1)

Application Number Title Priority Date Filing Date
CN201180004870.0A Expired - Fee Related CN103168298B (en) 2010-03-16 2011-03-07 Based on the virtual assistance that search engine is inferred

Country Status (5)

Country Link
US (4) US8214344B2 (en)
JP (1) JP5426779B2 (en)
KR (1) KR101356310B1 (en)
CN (2) CN103168298B (en)
WO (1) WO2011115768A2 (en)

Families Citing this family (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8484222B1 (en) 2006-12-01 2013-07-09 Google Inc. Method and apparatus for identifying a standalone location
US8214344B2 (en) 2010-03-16 2012-07-03 Empire Technology Development Llc Search engine inference based virtual assistance
DE102010032229A1 (en) * 2010-07-26 2012-01-26 Elektrobit Automotive Gmbh Technique for determining points of interest for a navigation device
US20120047247A1 (en) * 2010-08-18 2012-02-23 Openwave Systems Inc. System and method for allowing data traffic search
US20130273797A1 (en) * 2012-04-17 2013-10-17 E I Du Pont De Nemours And Company Conductive polymer layers grafted onto insulating polymer surfaces
US8984080B1 (en) 2013-04-09 2015-03-17 Kabam, Inc. Facilitating user configured assistance requests through a chat in a virtual space
US9332043B1 (en) 2013-04-12 2016-05-03 Kabam, Inc. System and method for facilitating user interaction with a virtual space through a graphical chat interface
WO2015178715A1 (en) * 2014-05-23 2015-11-26 Samsung Electronics Co., Ltd. System and method of providing voice-message call service
US9161193B1 (en) * 2014-06-04 2015-10-13 Grandios Technologies, Llc Advanced telephone management
WO2016081749A1 (en) 2014-11-19 2016-05-26 Google Inc. Methods, systems, and media for presenting related media content items
US10607609B2 (en) * 2016-08-12 2020-03-31 Magic Leap, Inc. Word flow annotation
KR101934280B1 (en) 2016-10-05 2019-01-03 현대자동차주식회사 Apparatus and method for analyzing speech meaning
US10740373B2 (en) * 2017-02-08 2020-08-11 International Business Machines Corporation Dialog mechanism responsive to query context
US10229683B2 (en) * 2017-03-10 2019-03-12 Soundhound, Inc. Speech-enabled system with domain disambiguation
US10956435B2 (en) * 2017-05-05 2021-03-23 Servicenow, Inc. Global search
DE102017118724A1 (en) * 2017-08-16 2019-02-21 Unify Patente Gmbh & Co. Kg Telecommunication terminal and method for controlling media streams
US11386131B2 (en) * 2018-05-29 2022-07-12 Microsoft Technology Licensing, Llc System and method for multi-language search
CN111161717B (en) 2019-12-26 2022-03-22 思必驰科技股份有限公司 Skill scheduling method and system for voice conversation platform

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6816858B1 (en) * 2000-03-31 2004-11-09 International Business Machines Corporation System, method and apparatus providing collateral information for a video/audio stream
CN1573924A (en) * 2003-06-20 2005-02-02 PtoPA株式会社 Speech recognition apparatus, speech recognition method, conversation control apparatus, conversation control method
CN1860544A (en) * 2003-09-25 2006-11-08 三星电子株式会社 Apparatus and method for displaying audio and video data, and storage medium recording thereon a program to execute the displaying method
CN1910654A (en) * 2004-01-20 2007-02-07 皇家飞利浦电子股份有限公司 Method and system for determining the topic of a conversation and obtaining and presenting related content
US20080005105A1 (en) * 2006-06-28 2008-01-03 Microsoft Corporation Visual and multi-dimensional search
CN101105810A (en) * 2003-06-07 2008-01-16 三星电子株式会社 Apparatus and method for displaying multimedia data

Family Cites Families (32)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
IL107482A (en) * 1992-11-04 1998-10-30 Conquest Software Inc Method for resolution of natural-language queries against full-text databases
US5844415A (en) 1994-02-03 1998-12-01 Massachusetts Institute Of Technology Method for three-dimensional positions, orientation and mass distribution
US6029195A (en) * 1994-11-29 2000-02-22 Herz; Frederick S. M. System for customized electronic identification of desirable objects
US5745899A (en) * 1996-08-09 1998-04-28 Digital Equipment Corporation Method for indexing information of a database
US6049796A (en) * 1997-02-24 2000-04-11 Nokia Mobile Phones Limited Personal digital assistant with real time search capability
US6101473A (en) * 1997-08-08 2000-08-08 Board Of Trustees, Leland Stanford Jr., University Using speech recognition to access the internet, including access via a telephone
US7107218B1 (en) * 1999-10-29 2006-09-12 British Telecommunications Public Limited Company Method and apparatus for processing queries
US7725307B2 (en) * 1999-11-12 2010-05-25 Phoenix Solutions, Inc. Query engine for processing voice based queries including semantic decoding
US6999932B1 (en) 2000-10-10 2006-02-14 Intel Corporation Language independent voice-based search system
JP4135307B2 (en) * 2000-10-17 2008-08-20 株式会社日立製作所 Voice interpretation service method and voice interpretation server
US20020087312A1 (en) * 2000-12-29 2002-07-04 Lee Victor Wai Leung Computer-implemented conversation buffering method and system
US6993535B2 (en) * 2001-06-18 2006-01-31 International Business Machines Corporation Business method and apparatus for employing induced multimedia classifiers based on unified representation of features reflecting disparate modalities
KR20030012284A (en) * 2001-07-31 2003-02-12 (주)한국음성도메인센터 Web browzer and controlling method thereof and web browzing service system
US20030115191A1 (en) * 2001-12-17 2003-06-19 Max Copperman Efficient and cost-effective content provider for customer relationship management (CRM) or other applications
US7398209B2 (en) * 2002-06-03 2008-07-08 Voicebox Technologies, Inc. Systems and methods for responding to natural language speech utterance
US7171351B2 (en) * 2002-09-19 2007-01-30 Microsoft Corporation Method and system for retrieving hint sentences using expanded queries
US7729913B1 (en) * 2003-03-18 2010-06-01 A9.Com, Inc. Generation and selection of voice recognition grammars for conducting database searches
BRPI0410362B1 (en) * 2003-05-16 2017-06-20 Google Inc. SYSTEMS AND METHODS OF SHARING NETWORK AND NETWORK MEDIA
US20050210042A1 (en) * 2004-03-22 2005-09-22 Goedken James F Methods and apparatus to search and analyze prior art
US7769705B1 (en) * 2004-05-21 2010-08-03 Ray Anthony Luechtefeld Method, artificially intelligent system and networked complex for facilitating group interactions
US7653627B2 (en) * 2005-05-13 2010-01-26 Microsoft Corporation System and method for utilizing the content of an online conversation to select advertising content and/or other relevant information for display
US10585942B2 (en) * 2005-09-14 2020-03-10 Millennial Media Llc Presentation of search results to mobile devices based on viewing history
JP4997743B2 (en) 2005-11-10 2012-08-08 日本電気株式会社 Document search apparatus, document search program, and document search method
US7739221B2 (en) * 2006-06-28 2010-06-15 Microsoft Corporation Visual and multi-dimensional search
US7711668B2 (en) * 2007-02-26 2010-05-04 Siemens Corporation Online document clustering using TFIDF and predefined time windows
JP2008293211A (en) 2007-05-23 2008-12-04 Hitachi Ltd Item recommendation system
JP2009070180A (en) * 2007-09-13 2009-04-02 Internatl Business Mach Corp <Ibm> Apparatus and method for supporting information retrieval
US7966304B2 (en) * 2007-11-30 2011-06-21 Yahoo! Inc. Enabling searching on abbreviated search terms via messaging
US8078632B1 (en) * 2008-02-15 2011-12-13 Google Inc. Iterated related item discovery
JP2010072507A (en) * 2008-09-22 2010-04-02 Toshiba Corp Speech recognition search system and speech recognition search method
CN101477556B (en) * 2009-01-22 2010-09-15 苏州智讯科技有限公司 Method for discovering hot spot in internet mass information
US8214344B2 (en) 2010-03-16 2012-07-03 Empire Technology Development Llc Search engine inference based virtual assistance

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6816858B1 (en) * 2000-03-31 2004-11-09 International Business Machines Corporation System, method and apparatus providing collateral information for a video/audio stream
CN101105810A (en) * 2003-06-07 2008-01-16 三星电子株式会社 Apparatus and method for displaying multimedia data
CN1573924A (en) * 2003-06-20 2005-02-02 PtoPA株式会社 Speech recognition apparatus, speech recognition method, conversation control apparatus, conversation control method
CN1860544A (en) * 2003-09-25 2006-11-08 三星电子株式会社 Apparatus and method for displaying audio and video data, and storage medium recording thereon a program to execute the displaying method
CN1910654A (en) * 2004-01-20 2007-02-07 皇家飞利浦电子股份有限公司 Method and system for determining the topic of a conversation and obtaining and presenting related content
US20080005105A1 (en) * 2006-06-28 2008-01-03 Microsoft Corporation Visual and multi-dimensional search

Also Published As

Publication number Publication date
CN103168298A (en) 2013-06-19
US10380206B2 (en) 2019-08-13
US8214344B2 (en) 2012-07-03
WO2011115768A3 (en) 2013-09-12
US20120233146A1 (en) 2012-09-13
JP2013515328A (en) 2013-05-02
US20130282692A1 (en) 2013-10-24
US8495051B2 (en) 2013-07-23
CN103168298B (en) 2016-03-09
KR101356310B1 (en) 2014-01-28
US20160004780A1 (en) 2016-01-07
KR20120085865A (en) 2012-08-01
US9201970B2 (en) 2015-12-01
WO2011115768A2 (en) 2011-09-22
JP5426779B2 (en) 2014-02-26
US20110231379A1 (en) 2011-09-22

Similar Documents

Publication Publication Date Title
CN105760464A (en) Method for automatically displaying inferences and computing device
CN109478205B (en) Architecture and method for computer learning and understanding
Lombardo et al. Storytelling on mobile devices for cultural heritage
EP2157571B1 (en) Automatic answering device, automatic answering system, conversation scenario editing device, conversation server, and automatic answering method
CN105051732B (en) The ranking of locally applied content
CN109155136A (en) The computerized system and method for highlight are detected and rendered automatically from video
CN107924679A (en) Delayed binding during inputting understanding processing in response selects
CN108027832A (en) The visualization of the autoabstract scaled using keyword
CN108920666A (en) Searching method, system, electronic equipment and storage medium based on semantic understanding
KR20160055930A (en) Systems and methods for actively composing content for use in continuous social communication
CN107808004A (en) Model training method and system, server, storage medium
EP3720060A1 (en) Apparatus and method for providing conversation topic
TW200900966A (en) Client input method
CN109036372A (en) A kind of voice broadcast method, apparatus and system
US20170364336A1 (en) Systems and methods for building conversational understanding systems
Albert Networked Communities: Strategies for Digital Collaboration: Strategies for Digital Collaboration
CN110377778A (en) Figure sort method, device and electronic equipment based on title figure correlation
CN110111611A (en) Information processing method, device, medium and electronic equipment
CN107046582A (en) Message prompt method, medium, device and the computing device of e-sourcing information
CN103488529B (en) A kind of method and apparatus for video resource access control
El Morr et al. Virtual community building and the information society: Current and future directions
Nigam et al. Towards a robust metric of polarity
CN108959450A (en) It is a kind of popular to comment on the system of determination, method, apparatus and readable medium
CN108351868A (en) The interactive content provided for document generates
CN106802941A (en) The generation method and equipment of a kind of reply message

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20160713

RJ01 Rejection of invention patent application after publication